Apple 发布在线强化学习框架，训练 UI 导航智能体直接在真实 iOS 应用上运行
Online Dense Reinforcement Learning with Planning-Grounding Agent for UI Navigation on Live iOS Apps
Apple 把 UI 导航智能体拆成规划和定位两个模块，用在线强化学习在真实 iOS 应用里训练，小模型也能自己纠错，不用每次都调大模型。
Apple 提出一个在线强化学习框架，让 UI 导航智能体直接在真实 iOS 应用上训练，而非依赖静态轨迹。该框架将智能体拆分为 Planning 和 Grounding 两个组件，分别负责决策与界面操作定位。相比 SFT 或离线强化学习训练的小模型容易记忆固定路径、出错后无法恢复，这一方法能在线上实时迭代并具备纠错能力。研究指出前沿视觉语言模型虽有 UI 导航能力，但推理成本过高，难以日常使用。
Online Dense Reinforcement Learning with Planning-Grounding Agent for UI Navigation on Live iOS Apps
While frontier vision-language models demonstrate strong capabilities in user interface (UI) navigation, their high inference costs prohibit everyday use. Conversely, smaller models trained via supervised fine-tuning (SFT) or offline reinforcement learning often memorize static trajectories, leading to brittle execution and an inability to recover from errors. To address these limitations, we propose a novel online reinforcement learning framework that trains UI navigation agents directly on live iOS applications. Our framework decouples the agent into two specialized components: a Planning…
